Introduction

The aliphatic terpolymer of ethylene, propylene and carbon monoxide (EPCO) has attracted much attention from polymer engineers and scientists because of its excellent physicochemical properties. Favorable properties such as high tensile strength, high wear resistance, low gas permeability and high chemical resistance give EPCO great potential as an engineering plastic and a barrier packaging material.1, 2, 3, 4, 5, 6, 7, 8, 9 However, control of the temperature to below 240°C is recommended for the melt fabrication process of EPCO because thermal degradation occurs readily at a higher temperature.10, 11, 12, 13, 14 This is a result of residual palladium, the polymerization catalyst, for EPCO, which frequently causes trouble in the melt processing of EPCO because of a very narrow processing window.1, 7 For example, the exceptionally high rate of crystallization in the injection molding process sometimes causes problems such as blockage of the sprue and warping of the resultant products.

It is recognized that most foreign materials lead to a higher crystallization rate of crystalline polymers. However, in addition to improving the notched impact properties, a polyester-based thermoplastic elastomer (HY) retards the crystallization of crystalline EPCO.15 In this study, we elucidated the retardation mechanism of the crystallization of EPCO by HY by investigating the crystallization behavior of EPCO in the static and dynamic states in terms of the dynamic shearing condition as well as blend composition.

Experimental Procedure

Materials

EPCO (Hyosung Co., Anyang, South Korea) and HY (Hytrel 4069, Dupont, USA) were vacuum dried at 60 °C for 12 h prior to use. EPCO was composed of randomly repeated units of -(CH2-CH2-C=O)- and -(CH2-CH(CH3)-C=O)-, the melt index of which was 60 at 240 °C and 2.16 kg load (ASTM D1238). HY was composed of poly(tetramethylene terephthalate) as a hard segment and poly(oxytetramethylene) glycol as a soft segment, as schematically shown in Scheme 1.16, 17 EPCO and HY were melt mixed in a Haake internal mixer at 240 °C at 80 rpm for 4 min. Blends containing 10, 20, 30 and 40 wt% HY were coded as HY10, HY20, HY30 and HY40, respectively.

Measurement of physical properties

Thermal properties were measured by differential scanning calorimetry (DSC Q20, TA Instruments, New Castle, DE, USA) in a nitrogen atmosphere. For differential scanning calorimetry measurement, the scan rate was 10 °C min−1 over a temperature range of −20 to 250 °C. To observe the isothermal crystallization process in the static state, the samples were held at 245 °C for 3 min in differential scanning calorimetry, then they were cooled to the preset temperature (Ti) to bring about isothermal crystallization.

Surface morphology of EPCO/HY blends was observed by scanning electron microscopy (FE-SEM, JSM-6340F, JEOL, Tokyo, Japan). The specimens were fractured in liquid nitrogen. The fractured surface was then etched by formic acid to dissolve the HY phase. The etched surface was sputter-coated with a thin Pt layer to avoid charging. The crystal structure of EPCO, HY and their blends was examined with a wide-angle X-ray diffractometer (WAXD, Rigaku Denki Co., Tokyo, Japan) with Nickel filtered CuKα radiation of 40 kV and 60 mA. A scanning experiment was carried out on the equator over the 2θ range of 5–50o at a scan speed of 7o min−1.

Dynamic mechanical properties of EPCO and HY were measured by dynamic mechanical analysis (DMA Q800, TA Instruments) at a rate of 5 °C min−1 with a load frequency of 1 Hz under a nitrogen atmosphere. Rheological properties were measured by an Advanced Rheometric Expansion System (TA Instruments) at 240 °C in a nitrogen atmosphere. A parallel plate geometry, the diameter and gap of which were 25 and 0.8 mm respectively, was adopted. The strain level was adjusted to 5% as indicated by preliminary tests. The Advanced Rheometric Expansion System specimen was prepared by compression molding to minimize residual stress. Prior to measurement, the melt was kept between the plates for 4 min to fully relax residual stresses. To observe the isothermal crystallization process under dynamic shear conditions, a time sweep measurement was carried out at 1 and 10 rad s−1 until the storage modulus (G') reached the ceiling value.

Results and Discussion

The relative crystallinity (Xt) with crystallization time (t) is calculated by Equation (1)18

in which dH/dt is the rate of heat evolution and t0 and t∞ are the times at which crystallization process begins and ends, respectively. Figure 1 shows the Xt of EPCO and EPCO/HY blends versus time at 197.5 °C. The half-time of crystallization (t0.5) of EPCO and the blends at the isothermal crystallization temperature (Ti) obtained from Figure 1 is given in Table 1. The t0.5 of the blends increases with increasing HY content. This indicates that HY reduces the crystallization rate of EPCO.

Figure 1
figure 1

Plots of relative crystallinity versus time for the quiescent isothermal crystallization of EPCO and EPCO/HY blends at 197.5 °C.

Full size image
Table 1 The half-time of crystallization and the Avrami parameters of EPCO and EPCO/HY blends at various isothermal temperatures
Full size table

The cooling and subsequent heating differential scanning calorimetry scans of EPCO, HY and their blends are shown in Figures 2a and b, respectively. As a rule, the incorporation of a foreign material increases the crystallization rate of a crystalline polymer by causing heterogeneous nucleation because the foreign material mostly has a role as a nucleating agent. Hence, heterogeneous crystallization at an interface leads to an increase in the Tc of a blend.18, 19, 20, 21, 22 As reported, HY is a foreign material because it has little miscibility with EPCO.15 This can be confirmed by the wide-angle X-ray diffractometer patterns of the neat polymers and their blends. In Figure 3, EPCO obviously shows strong peaks of β forms at 22, 25 and 28°, representing the (110), (111) and (210) planes, respectively.23 The blends show the unique EPCO peaks as well at the same scattering angle without generating new diffraction peaks. Therefore, the accelerated crystallization of EPCO is expected in principle. However, in reality, HY decelerates the crystallization of EPCO as evidenced by a decrease in the Tc of the immiscible blends with increasing HY content. This suggests that crystallization as well as nucleation of the blends occurs only in the neat EPCO phase without any contribution from the amorphous HY phase.24 However, the solvent-etched morphology of the blends in Figure 4 shows that the HY domains are widely dispersed in the form of droplet-like domains in the EPCO matrix. As a result, these immiscible HY domains seem to physically disturb the crystal growth of EPCO along the path of spherulite development. This results in a decreased crystal size and perfectness, which corresponds to the reduction of the Tm of the blends as shown in Figure 2b.

Figure 2
figure 2

DSC (a) cooling and (b) heating scan thermograms of EPCO, HY and EPCO/HY blends.

Full size image
Figure 3
figure 3

WAXD profiles of EPCO, HY and EPCO/HY blends.

Full size image
Figure 4
figure 4

The solvent etched morphology of (a) HY10, (b) HY20, (c) HY30 and (d) HY40, respectively. HY was extracted with formic acid.

Full size image

Figure 5 shows the Avrami plots of EPCO and its blends at 197.5 °C. The Avrami exponent (n) and crystallization rate constant (Kt) obtained at different Tis are listed in Table 1. The n is a measure of the dimensionality of the spherulite growth and the average values for each sample at several Ti decrease from 2.08 to 1.68 with an increasing HY content, which indicates the reduced dimensionality of the EPCO spherulites. In addition, as the HY content increases the Kt values decrease, which further verifies the retardation of the crystallization of EPCO.

Figure 5
figure 5

Avrami plots for the quiescent isothermal crystallization of EPCO and EPCO/HY blends at 197.5 °C.

Full size image

The variation of storage moduli (G') of two polymers with temperature is plotted in Figure 6. HY gives much lower values of G' than EPCO over the temperature range shown, indicating a very low melt viscosity of HY. Furthermore, the phase character of HY and EPCO at Tc can be verified by the variation of loss tangent (tanδ) in Figure 7. A greater value of tanδ indicates a higher liquidity of the samples.25 HY gives greater tanδ than EPCO above 188 °C, Tc of EPCO, which means a higher liquid character of the HY phase than for the EPCO matrix over the temperature range of crystallization. Hence, the liquid-like HY phase easily shields the EPCO spherulites during the crystal growth, as illustrated in Scheme 2.26, 27 In addition, the two components may be slightly miscible to a limited extent because both polymers contain carbonyl groups in the main chain. As a consequence, it is expected that the crystal growth of EPCO in the blend was retarded because the EPCO chains have to displace the HY moiety from the spherulite surface during the crystal growth process.

Figure 6
figure 6

The variation of the storage modulus (G') of EPCO and HY with increasing temperature. The marked temperature on the graph is the Tc of EPCO.

Full size image
Figure 7
figure 7

The variation of tanδ for EPCO and HY with increasing temperature. The marked temperature on the graph is the Tc of EPCO.

Full size image

The effect of the shear stress on the crystallization of polymeric systems is evaluated by dynamic rheological measurement at a given temperature. This can be done by tracing the variation of G' with time at the temperature. Figure 8 shows the variation of the G' of the EPCO and EPCO/HY blends with time at different frequencies at 197 °C. G' increases very slowly during the early stages. This period is referred as the induction time for crystallization. The induction period is defined as the stage when randomly entangled polymer chains transform to a regularly aligned lattice. Then, an abrupt increase of G' follows because the homogeneous melt becomes a heterogeneous system with the formation and growth of crystallites. The induction time of EPCO and its blends at different frequencies are given in Table 2. The induction time is notably increased with an increasing HY content at all frequencies considered, indicating that HY still reduces the crystallization rate of EPCO even in the dynamic state.

Figure 8
figure 8

Shear-induced isothermal crystallization of EPCO and EPCO/HY blends at 197 °C and (a) 1 rad s−1 and (b) 10 rad s−1, respectively.

Full size image
Table 2 The induction time of EPCO and EPCO/HY blends under the dynamic condition at 197oC
Full size table

It is noteworthy that the induction time of the blends is slightly decreased with an increase in the shear rate. In most cases, shear-induced crystallization takes place under dynamic conditions in polymer systems.18, 19, 21 However, when shear stress is applied to the EPCO/HY blends, it may be almost concentrated on the easily deformable HY domains because of the lower viscosity and Newtonian character, as previously mentioned. Then, the HY phase rapidly dissipates the large stresses. The cavity on the interface is direct evidence for the stress-dissipation of the HY elastomer.15 Hence, the shear rate has little effect on the crystallization behavior of the EPCO/HY blends.

Conclusion

The crystallization behavior of the EPCO/HY blends in the static and dynamic states revealed that HY retarded the crystallization of EPCO without changing its crystal structure. The large difference in the elastic moduli of the two polymers at the crystallization temperature was the principal reason for the retardation process. This is a somewhat unexpected result. However, this result may be effectively utilized in the design of injection molding processes for EPCO by widening the processing window.

scheme 1

The chemical structure of Hytrel.

scheme 2

The retardation mechanism of the crystallization of EPCO by the incorporation of HY in the blend phase. A full color version of this figure is available at Polymer Journal online.